Children with peer relationship problems have been shown to be poorly adjusted and risk subsequent problems in mental health and social adjustment. Social skills training, social cognitive interventions, and co-operative group programs are promising approaches for treating these problems in the classroom. These approaches, however, have several limitations: a) the evaluation of programs has often been limited, particularly in terms of observational measures, the assessment of implementation, and the determination of whether effects are maintained; b) studies have not determined whether classroom-based programs affect children with peer relationship problems as well as those without; and c) the programs are usually implemented by research investigators or a select group of teachers. It is unclear if ordinary teachers can or will implement such programs such programs and if they do, how well they can implement them and whether changes will be effected in themselves or their students. In the proposed study, teachers will be provided workshops and consultation on peer intervention programs. One school will receive the consultation program during the first and second year, whereas a second school will serve as a control and not begin to receive it until the second year. Data will be collected at the beginning and end of the first two years and at the beginning of the third. The evaluation includes measures of: a) principals' perceptions of teachers, b) teachers' attitudes toward and use of peer intervention programs, c) teacher ratings of children's social skills and individual problems, d) children's perceptions of peer relations, school life, and their own competence, e) sociometric measures, and f) children's classroom interactions. Measures of implantation (level of usage, innovation configurations, and stage of concern) will be collected every six weeks. Although the primary focus is on stage of implantation, the study address several methodological and theoretical issues as well.